Papers by Michael Golosovsky
La2/3Sr1/3MnO3薄膜における強磁性不均一性の研究
Journal of Physics D, 2017
We discuss microscopic mechanisms of complex network growth, with the special emphasis of how the... more We discuss microscopic mechanisms of complex network growth, with the special emphasis of how these mechanisms can be evaluated from the measurements on real networks. As an example we consider the network of citations to scientific papers. Contrary to common belief that its growth is determined by the linear preferential attachment, our microscopic measurements show that it is driven by the nonlinear autocatalytic growth. This invalidates the scale-free hypothesis for the citation network. The nonlinearity is responsible for a dramatic dynamical phase transition: while the citation lifetime of majority of papers is 6-10 years, the highly-cited papers have practically infinite lifetime. Keywords power-law distribution • citations • preferential attachment • complex networks • autocatalytic growth
Cornell University - arXiv, Feb 24, 2020
Universality or near-universality of citation distributions was found empirically a decade ago bu... more Universality or near-universality of citation distributions was found empirically a decade ago but its theoretical justification has been lacking so far. Here, we systematically study citation distributions for different disciplines in order to characterize this putative universality and to understand it theoretically. Using our calibrated model of citation dynamics, we find microscopic explanation of the universality of citation distributions and explain deviations therefrom. We demonstrate that citation count of the paper is determined, on the one hand, by its fitness-the attribute which, for most papers, is set at the moment of publication. The fitness distributions for different disciplines are very similar and can be approximated by the log-normal distribution. On another hand, citation dynamics of a paper is related to the mechanism by which the knowledge about it spreads in the scientific community. This viral propagation is non-universal and disciplinespecific. Thus, universality of citation distributions traces its origin to the fitness distribution, while deviations from universality are associated with the discipline-specific citation dynamics of papers.
Statistical distributions with heavy tails are ubiquitous in natural and social phenomena. Since ... more Statistical distributions with heavy tails are ubiquitous in natural and social phenomena. Since the entries in heavy tail have unproportional significance, the knowledge of its exact shape is very important. Citations of scientific papers form one of the best-known heavy tail distributions. Even in this case there is a considerable debate whether citation distribution follows the log-normal or power-law fit. The goal of our study is to solve this debate by measuring citation distribution for a very large and homogeneous data. We measured citation distribution for 418,438 Physics papers published in 1980-1989 and cited by 2008. While the log-normal fit deviates too strong from the data, the discrete power-law function with the exponent γ = 3.15 does better and fits 99.955% of the data. However, the extreme tail of the distribution deviates upward even from the power-law fit and exhibits a dramatic "runaway" behavior. The onset of the runaway regime is revealed macroscopically as the paper garners 1000-1500 citations, however the microscopic measurements of autocorrelation in citation rates are able to predict this behavior in advance.
Your article is protected by copyright and all rights are held exclusively by Springer Science +B... more Your article is protected by copyright and all rights are held exclusively by Springer Science +Business Media New York. This e-offprint is for personal use only and shall not be self-archived in electronic repositories. If you wish to self-archive your work, please use the accepted author’s version for posting to your own website or your institution’s repository. You may further deposit the accepted author’s version on a funder’s repository at a funder’s request, provided it is not made publicly available until 12 months after publication.
Stochastic Modeling of References and Citations
SpringerBriefs in Complexity, 2019
We discuss here how the authors compose reference lists of their papers. We assume that the autho... more We discuss here how the authors compose reference lists of their papers. We assume that the author chooses his references basing on the copying/recursive search algorithm. We put this algorithm at the core of the quantitative model accounting for the age composition of the reference lists of papers. The model contains two empirical functions: the aging function and the obsolescence function, and one empirical parameter—paper’s fitness. At the next step, we calibrate this model in measurements with Physics papers, namely, we determine the aging function, obsolescence function, and paper’s fitness averaged over the groups of similar papers. At the next step, we extend our approach to individual papers and formulate a probabilistic model of their citation dynamics. This model represents citation dynamic of a paper as a self-exciting stochastic process (Hawkes process). We demonstrate that the mean-field approximation of this probabilistic model yields the age distribution of references, as expected.
Complex Network of Scientific Papers
SpringerBriefs in Complexity, 2019
We consider scientific publications as a growing complex network where papers are nodes and citat... more We consider scientific publications as a growing complex network where papers are nodes and citations are links which connect these papers together. We explain this network approach and make special accent on the temporal aspect of citation network, namely, we focus on the growth of the number of papers, age distribution of references, and citation dynamics. We trace relation of the age distribution of references to citation dynamics and explore this reference-citation duality.
Microscope a resistivite en champ proche
La presente invention concerne un microscope hyperfrequence comportant un guide d'ondes hyper... more La presente invention concerne un microscope hyperfrequence comportant un guide d'ondes hyperfrequence pourvu d'une extremite de sonde se placant a proximite de la surface a sonder, pour que la surface se trouve dans le champ proche du rayonnement hyperfrequence. L'extremite de la sonde regardant vers la surface est recouverte d'une feuille metallique ou d'un film electroconducteur depose sur une couche dielectrique dans lesquels est pratiquee une fente rectangulaire. La longueur de la fente est proche de la longueur de resonance de l'onde hyperfrequence, cette derniere etant juste legerement superieure a une demi-longueur de l'onde hyperfrequence. La largeur, qui est sensiblement plus petite que la longueur, est selectionnee de facon que l'extremite portant la fente soit transparente a l'onde hyperfrequence. Il en resulte que l'importante emission d'energie hyperfrequence par la fente se fait avec une finesse de resolution qui est fonctio...
Author response for "Uncited papers are not useless
Comparison to Existing Models
SpringerBriefs in Complexity, 2019
We make a survey of models of citation dynamics and focus on the preferential attachment and fitn... more We make a survey of models of citation dynamics and focus on the preferential attachment and fitness models. We show that under certain realistic conditions these models are equivalent. In order to find the microscopic foundations of the preferential attachment mechanism, we analyze theoretically and experimentally several citation networks and demonstrate that, for a broad fitness distribution, this mechanism reduces to the fitness model. The fitness model yields the long-sought explanation for the initial attractivity K0, an elusive parameter which was left unexplained within the framework of the empirical preferential attachment model. We show that the initial attractivity is determined by the width of the fitness distribution. We compare the preferential attachment and fitness models to our microscopic model of citation dynamics based on recursive search and show that our model contains both these phenomenological models.
ArXiv, 2020
We study citation dynamics of the Physics, Economics, and Mathematics papers published in 1984 an... more We study citation dynamics of the Physics, Economics, and Mathematics papers published in 1984 and focus on the fraction of uncited papers in these three collections. Our model of citation dynamics, which considers citation process as an inhomogeneous Poisson process, captures this uncitedness ratio fairly well. It should be noted that all parameters and variables in our model are related to citations and their dynamics, while uncited papers appear as a byproduct of the citation process and this is the Poisson statistics which makes the cited and uncited papers inseparable. This indicates that the most part of uncited papers constitute the inherent part of the scientific enterprise, namely, uncited papers are not unread.
Citation Dynamics of Individual Papers: Model Calibration
SpringerBriefs in Complexity, 2019
The model of citation dynamics developed in Chap. 3 is compared to measurements in order to deter... more The model of citation dynamics developed in Chap. 3 is compared to measurements in order to determine empirical functions and parameters such as aging function, obsolescence function, and the paper’s fitness. We found that the aging function is universal, namely, it is the same for all papers in one field published in the same year. However, the obsolescence function depends on the number of previous citations. This unexpected finding prompted us to focus more closely on the network aspect of citation dynamics and to consider not only the nearest neighbors of each paper in the citation network, but its next-nearest neighbors as well. The updated model takes into account the correlations between citation dynamics of a paper and its neighbors (network assortativity).
We studied the ∞uorescence emission of the Enhanced Green Fluorescent Protein (EGFP) in aqueous s... more We studied the ∞uorescence emission of the Enhanced Green Fluorescent Protein (EGFP) in aqueous solution under continuous microwave irradiation with a well-deflned fleld pattern (a TE011 microwave cavity operating at 9.5GHz). We focused on polarization of the emitted light and measured the spectrum of the ∞uorescence emission anisotropy. We found that in the spectral range 500{540nm the microwave-induced efiect can be reduced to heating, while in the spectral range 540{560nm, the microwave-induced efiect difiers from that resulting from conventional heating.
ArXiv, 2018
We analyze the growth models for complex networks including preferential attachment (A.-L. Baraba... more We analyze the growth models for complex networks including preferential attachment (A.-L. Barabasi and R. Albert, Science 286, 509 (1999)) and fitness model (Caldarelli et al., Phys. Rev. Lett. 89, 258702 (2002)) and demonstrate that, under very general conditions, these two models yield the same dynamic equation of network growth, dK dt = A(t)(K +K0), where A(t) is the aging constant, K is the node’s degree, and K0 is the initial attractivity. Basing on this result, we show that the fitness model provides an underlying microscopic basis for the preferential attachment mechanism. This approach yields long-sought explanation for the initial attractivity, an elusive parameter which was left unexplained within the framework of the preferential attachment model. We show that K0 is mainly determined by the width of the fitness distribution. The measurements of K0 in many complex networks usually yield the same K0 ∼ 1. This empirical universality can be traced to frequently occurring log...
Quantitative Science Studies, 2021
Universality of scaled citation distributions was claimed a decade ago but its theoretical justif... more Universality of scaled citation distributions was claimed a decade ago but its theoretical justification has been lacking so far. Here, we study citation distributions for three disciplines—Physics, Economics, and Mathematics—and assess them using our explanatory model of citation dynamics. The model posits that the citation count of a paper is determined by its fitness: the attribute, which, for most papers, is set at the moment of publication. In addition, the papers’ citation count is related to the process by which the knowledge about this paper propagates in the scientific community. Our measurements indicate that the fitness distribution for different disciplines is nearly identical and can be approximated by the log-normal distribution, while the viral propagation process is discipline specific. The model explains which sets of citation distributions can be scaled and which cannot. In particular, we show that the near-universal shape of the citation distributions for differen...
Quantitative Science Studies, 2021
We study the citation dynamics of the papers published in three scientific disciplines (Physics, ... more We study the citation dynamics of the papers published in three scientific disciplines (Physics, Economics, and Mathematics) and four broad scientific categories (Medical, Natural, Social Sciences, and Arts & Humanities). We measure the uncitedness ratio, namely, the fraction of uncited papers in these data sets and its dependence on the time following publication. These measurements are compared with a model of citation dynamics that considers acquiring citations as an inhomogeneous Poisson process. Themodel captures the fraction of uncited papers in our collections fairly well, suggesting that uncitedness isan inevitable consequence of the Poisson statistics.
Citation Analysis and Dynamics of Citation Networks, 2019
We apply stochastic model of citation dynamics of individual papers developed in our previous wor... more We apply stochastic model of citation dynamics of individual papers developed in our previous work (M. Golosovsky and S. Solomon, Phys. Rev. E 95, 012324 (2017)) to forecast citation career of individual papers. We focus not only on the estimate of the future citations of a paper but on the probabilistic margins of such estimate as well.
Le Journal de Physique Colloques, 1981
Model Validation
SpringerBriefs in Complexity, 2019
Physical Review E, 2018
We analyze the growth models for complex networks including preferential attachment (A.-L.
Uploads
Papers by Michael Golosovsky